Cancer therapies often kill tumors by making chromosome breaks, and exceeding the tumor cell's ability to repair them. Nonhomologous end joining (NHEJ) resolves the majority of such breaks, and it has been suggested that cancers might differ significantly in how well they perform NHEJ. We will determine how much NHEJ fitness varies in a variety of cancers. We will measure efficiency of NHEJ with a specific and sensitive quantitative polymerase chain reaction (qPCR) assay. We will additionally employ next generation sequencing technology to characterize in parallel NHEJ accuracy. Importantly, cancer therapies that introduce chromosome breaks are effective in killing cells because they introduce breaks that are dirty - the break often has damage to DNA flanking the break that makes it difficult to repair. Using cell lines with known NHEJ defects (including cells from patients with inherited NHEJ defects), we have shown the employment of dirty end structures is critical if NHEJ is to evaluated in a manner consistent with the ability of cells, both cancer and surrounding normal cells, to survive cancer therapies. We will systematically profile NHEJ fitness in both a panel of cancer cell lines as well as in primary cancers. We will determine the extent of cancer-associated variation in NHEJ, and whether this variation could help predict therapy outcome and aid in development of better (rationally designed and targeted) therapies.